The shielding effect of wild type iron reducing bacterial floraon the corrosion of linepipe steel

Faisal M. AlAbbas a,!, Shaily M. Bhola a, John R. Spear b, David L. Olson a, Brajendra Mishra a

a Department of Metallurgical and Materials Engineering, Colorado School of Mines, Golden, CO 80401, USAb Department of Civil and Environmental Engineering, Colorado School of Mines, Golden, CO 80401, USA

a r t i c l e i n f o

Article history:Received 12 January 2013Received in revised form 4 May 2013Accepted 28 May 2013Available online 7 June 2013

Keywords:CorrosionImpedanceIron reducing bacteriaIRBBiofilm

a b s t r a c t

Microbiologically influenced corrosion (MIC) by microbes capable of iron reduction (ironreducing bacteria (IRB)) on API 5L !52 carbon steel coupons was investigated. A wild typeof IRB was isolated and cultivated from a water sample collected from a sour oil welllocated in Louisiana, USA. 16S rRNA gene sequence analysis indicated that the mixed bac-terial consortium contained two phylotypes close to members of the Proteobacteria (Shewa-nella oneidensis sp.) and Firmicutes (Brevibacillus sp.). The corrosion behavior of carbon steelcoupons exposed to different media, with and without these microbes, was characterizedby open circuit potential (OCP), electrochemical impedance spectroscopy (EIS) and polari-zation resistance (Rp), and a corrosion mechanism has been proposed. The biofilm and pitmorphology that developed with time were characterized using field emission scanningelectron microscopy (FESEM). Interestingly, surface morphology and electrochemical eval-uations confirmed that IRB metabolic activities and resulting biofilms inhibit the corrosionprocess. The maximum corrosion rate in the biotic system was 4 mpy, while it was 20 mpyin the abiotic solution. Minor isolated pits were revealed in the biotic system, whereasextensive general pitting was found in the abiotic system. Elemental analysis and corrosionproduct structures were characterized by energy-dispersive X-ray spectroscopy (EDS) andX-ray diffraction (XRD). XRD confirmed the formation of a significant amount of iron oxidecompounds that include iron, Hematite (Fe2O3), Magnetite (Fe3O4) and iron (II) hydroxideFe(OH)2 on the steel surface exposed to a biotic system.

! 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Carbon steel pipelines are considered the most efficient and economical method to transport hydrocarbon in the oil andgas industry. During oil and gas operations, pipeline networks are subjected to different corrosion deterioration mechanismsthat result from the interaction between the fluid process and steel. Among these corrosion mechanisms is microbiologicallyinfluenced corrosion (MIC). MIC is believed to be responsible for greater than 20% of the total corrosion cost in the oil and gasindustry [1]. Microorganisms are capable of causing catastrophic failures in oil and gas structures including pipeline systems,storage terminals and refineries [2]. Since the advent of modern oil and gas production, scientists and engineers have facedproblems caused by microorganisms. Indigenous microorganisms that naturally reside in hydrocarbon feeds and associatedsecondary systems are able to induce localized changes in the aqueous environment (e.g., alter the concentration of theelectrolyte components pH and oxygen concentration) leading to the localized forms of corrosion such as pitting and creviceformation [1,3].

1350-6307/$ - see front matter ! 2013 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.engfailanal.2013.05.020

! Corresponding author. Tel.: +1 720 532 4110; fax: +1 303 273 3795.E-mail address: falabbas@mines.edu (F.M. AlAbbas).

Engineering Failure Analysis 33 (2013) 222–235

Contents lists available at SciVerse ScienceDirect

Engineering Failure Analysis

journal homepage: www.elsevier .com/locate /engfai lanal

Petroleum hydrocarbons are an excellent carbon (food) source for a wide variety of microbes in all three domains of life—the Bacteria, Archaea and Eucarya—and microbial representatives of all three domains likely play roles in MIC [3,4]. The maintypes of microorganisms associated with metals corrosion in pipeline systems are sulfate reducing bacteria (SRB), iron reduc-ing bacteria (IRB), iron and manganese oxidizing bacteria (I/MOB), acid producing bacteria (APB) and methanogensis [3,4].SRB have been cited as the major MIC causative bacteria in the oil and gas industry [1,3,4]. In practical situations, MIC resultsfrom synergistic interactions of different microorganisms and consortia, which coexist in the environment and are able toaffect the chemical and electrochemical processes through co-operative metabolisms.

Unfortunately, the focus on SRB in MIC studies underestimates the role of other types of microorganisms (i.e., IRB) in thecorrosion process [1]. As early as 1980, Westlake and colleagues were able to cultivate and characterize Pseudomonas ferri-reductans from different oilfield samples. This type of bacteria is capable of reducing iron and has been referred to as IRB[5].IRB are facultative anaerobic where they use oxygen under aerobic conditions and ferric ion (Fe+3) reduction under anaerobicconditions [1,3,4]. It has been reported that IRB are able to metabolize other different electron acceptors including Mn(IV),NO3", NO2", S2O3

2", SO3 [1,3–7]Although the role of some types of microorganisms (i.e., SRB) is well established in MIC investigations, the involvement of

IRB in MIC is still debatable [6]. Conflicting investigations have reported both accelerating and inhibiting actions of IRB to-wards the corrosion process [6,8,9]. It has been postulated that IRB accelerate the corrosion by the following actions: (1) thereduction of insoluble ferric ion compounds to soluble ferrous ion that, in turn, remove the protective corrosion scalesformed on exposed surfaces and (2) promote the formation of concentration cells among the biofilm [10]. Corrosion effectsof pseudomonas sp. of mild steel were reported under microaerobic conditions, which were attributed to anodic polarizationdue to the ability of IRB to remove the protective ferric compounds [11]. Javaherdashti et al. [12] demonstrated that mildsteel exposed to a culture of IRB fails faster than under abiotic conditions, implying the corrosion enhancement action ofIRB. Conversely, there is evidence that IRB could inhibit the corrosion process under specific conditions. IRB could inhibitthe corrosion process via different mechanisms including: (1) aerobic respiration by removing the oxygen from the system[8,13] and (2) the presence of IRB along with SRB in the biofilm could be beneficiary as they could destroy the ecologicalnests of sulfate-reducers within the biofilms formed on metal surfaces [14]. Little et al. [15] reported short-term protectionof biofilms composed of a mixed culture of IRB and SRB. However, there are no certainties under real-world situationsregarding whether the IRB will inhibit or promote the corrosion. The complex nature of the biofilms developed on the cor-roding material, diverse microorganism structure and induced multiple chemical and electrochemical interactions result inthese uncertainties. The corrosion mechanisms and subsequent rates will depend on conditions of interfaces (solution/bio-film/surface) and the activities of sessile bacteria prevailing in the biofilm [14].

The goal of this investigation is to study the impact of an environmentally wild type of aerobic bacteria (cultivated fromoil field samples rather than obtained from a culture collection) on the corrosion behavior of API 5L !52 carbon steel. Theaerobic bacteria consortium used in this study was cultivated from a sour oil well in Louisiana, USA at 2200 ft depth.

2. Materials and methods

2.1. Organisms and testing medium

The aerobic bacteria consortium used in this study was cultivated from water samples obtained from a sour oil well lo-cated in Louisiana, USA. The water samples were collected and bottled at the wellhead from an approximate depth of 2200 ft.as described by NACE Standard TM0194 [16]. For IRB isolation, one milliliter of the water sample was transferred to a 250 mlof Erlenmeyer flask containing 100 ml of nutrient enriched solution that was composed of tryptone (10 g), sodium chloride(10 g), and yeast extract (5.0 g) added to one liter of distilled water [15]. The pH of the medium was adjusted to 7.2 using 5 Msodium hydroxide and autoclaved at 121 "C for 20 min. The bacteria were incubated at 30 "C in a rotary shaker at 150 rpmuntil turbid growth was observed.

2.2. Identification of the iron-reducing consortium

DNA was extracted from cultivars using the MoBio Power soil DNA extraction kit. 16S rRNA gene amplification was car-ried out using the ‘universal’ polymerase chain reaction (PCR) primers 515F and 1391R. PCR, cloning and transformationwere then carried out as described by Sahl et al. [17]. Unique restriction fragment length polymorphisms (RFLP) were se-quenced on an ABI 3730 DNA sequencer at Davis Sequencing, Inc. (Davis, CA). Sanger reads were called with PHRED viaXplorseq [18]. Sequences were compared to the GenBank database via BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi).

2.3. Sample Preparation

Pipeline steel (API 5L !52) coupons, provided by a local energy company, were used for this study, and there chemicalcompositions are shown in Table 1. Metallographic specimens from the received materials were prepared with standardmethods for optical microscopy (1 lm final polish and 2% nital etch). Representative microstructures are shown in Fig. 1,and it can be seen that API 5L !52 carbon steel contains a mixture of polygonal ferrite and pearlite structures.

F.M. AlAbbas et al. / Engineering Failure Analysis 33 (2013) 222–235 223

For corrosion evaluations, the coupons were machined to a size of 10 mm ! 10 mm ! 5 mm and embedded in a mold ofnon-conducting epoxy resin, leaving an exposed surface area of 100 mm2. For electrical connection, a copper wire was sol-dered at the rear of the coupons. The coupons were polished with a progressively finer sand grinding paper until a final gritsize of 600 lm was obtained. After polishing, the coupons were rinsed with distilled water, ultrasonically degreased in ace-tone and sterilized by exposing them to pure ethanol for 24 h.

2.4. Electrochemical tests

Electrochemical measurements were made in a conventional three electrode ASTM cell coupled with a potentiostat and ahigh frequency impedance analyzer as illustrated in Fig. 2. The electrochemical cells were composed of a test coupon as aworking electrode (WE), a graphite electrode as an auxiliary electrode and a saturated calomel electrode (SCE) as a reference.The glasswares were autoclaved at 121 "C for 20 min and air dried. Graphite electrodes, purging tubes, rubber stoppers andneedles were sterilized by immersion in 70 vol.% ethanol for 24 h followed by exposure to a UV lamp for 20 min. Two solu-tions were used in this experiment. Under sterilized conditions (in a sterilized laminar flow hood), the first cell was preparedwith 700 ml of sterilized nutrient enriched solution (described above) and the second cell was prepared with 700 ml

Table 1The chemical composition of API 5L !52 carbon steel coupons (weight%).

Fe C Si Cr Ni Mn Cu Mo Nb Ti Al V S P

Bal. 0.070 0.195 0.03 0.02 1.05 0.05 0.004 0.021 0.001 0.029 0.003 0.008 0.008

Fig. 1. Light microscopy for API 5L !52 carbon steel etched with 2 pct. nital etch showing microstructural features.

Fig. 2. A schematic illustration of the electrochemical setup.

224 F.M. AlAbbas et al. / Engineering Failure Analysis 33 (2013) 222–235

sterilized solution inoculated with 5 ml of the aerobic consortium at 106 cells/ml. To simulate the flow in the pipeline, elec-trochemical tests were performed at a flow rate of 0.9 m/s (100 rpm) under atmospheric pressure conditions at 30 "C.

Open circuit potential values of the systems were monitored with time during the immersion period followed by periodicreadings up to 336 h. Impedance measurements were performed on the system of the open circuit potential for various timeintervals from immersion up to 288 h. The frequency sweep was applied from 105 to 10"2 Hz with an AC amplitude of 10 mV.The linear polarization resistance (LPR) was measured on the system at a scanning amplitude of ±10 mV with reference tothe open circuit potential for various time intervals from immersion up to 336 h.

2.5. Surface analysis of the coupons exposed to IRB

At the conclusion of each test, the working electrodes were carefully removed from the system for fixation. To fix thegrown biofilm to the steel surface, the coupons were immersed for 1 h in a 2% glutaraldehyde solution, dehydrated with25, 50, 75 and 100 vol.% ethanol solutions (for 15 min each) successively, air dried overnight and then gold sputtered. Afterfixation, the coupons were examined via field emission scanning electron microscopy (FESEM) coupled with energy disper-sive spectroscopy (EDS). Corrosion products composition was obtained using the X-ray diffraction (XRD) method with a Phi-lips PW 3040/60 spectrometer using Cu Ka radiation source. The coupons were then cleaned following the ASTM-GI-0313[19] procedure, and the pit morphology and density on the exposed coupons were examined by FESEM.

3. Results and discussion

3.1. Identification of the aerobic bacteria consortium

Based on 99.9% similarity, 16S rRNA gene analysis indicated that the mixed bacterial culture consortium contained twophylotypes that have been related to members of the Proteobacteria (Shewanella oneidensis sp.) and Firmicutes (Brevibacillussp.). S. oneidensis are facultative anaerobic IRB that are capable of diverse metabolisms. They are able to reduce ferric iron andsulfite, oxidize hydrogen gas, and produce sulfide [20]. It has been reported that these bacteria might be involved in biocor-rosion [20], whereas some studies suggest that S. oneidensis may have an inhibitory effect towards corrosion [8,19].

On the other hand, Brevibacillus are aerobic spore-forming Firmicutes microbes. They have the ability to degrade hydro-carbons and plastics [21]. They were used to treat part of the Daqing oil field, in which the oil production increased by 165%for about 200 days [21]. Laboratory investigation has revealed that Brevibacillus provide stable operations in highly compe-tent microbial fuel cells [22]. Pham et al. [23] have demonstrated that Brevibacillus can achieve extracellular electron transferin the presence of Pseudomonas sp. This complimentary bacterial interaction is considered to be an instrumental process inthe anodic electron transfer of the microbial fuel cell (MFC) [23].

3.2. Morphology and composition of interfacial surfaces after exposure

The morphology observations of corrosion products of API 5L !52 carbon steel exposed to a sterilized control medium(abiotic) and inoculated medium (biotic) are shown in Fig. 3a and b, respectively. There are significant differences in thecharacteristics, composition and structures of the layers developed in the presence of the IRB microbial consortium (biotic)compared to the abiotic system. For the abiotic system, there is one homogenous layer of corrosion products (Fig. 3a) formedon the surface with some deposited salt crystals. Quantitative elemental EDS analysis shown in Fig. 3c and Table 2, revealedpeaks for oxygen, iron, sodium, chloride and carbon that accumulated from the growth medium. The EDS elemental analysissuggests that the corrosion layer might be composed mainly of a mixture of iron oxides, iron chlorides, sodium chlorides andorganic compounds.

In comparison, the significant number of products observed in the biotic system, Fig. 3b, is most likely due to the produc-tion of a biofilm matrix. The bacterial growth in the biotic system was evidenced by the solution turbidity and unpleasantdead fish smell. In the case of the biotic conditions, the EDS analysis, Fig. 3d and Table 2, revealed peaks for nitrogen andphosphorus in addition to oxygen, iron, sodium, chloride and carbon. The presence of the nitrogen and phosphorus is attrib-uted to the microbial production of extracellular polymer substance (EPS) as well as Fe-EPS complex [24]. The EDS resultssuggest that in the biotic system, the steel surface is covered with a mixed layer of corrosion products and biofilm matrix.

The structure of the biofilm developed by the presence of the IRB consortium together with the produced corrosion prod-ucts is shown in Fig. 4. The FESEM micrograph (Fig. 4) reveals the presence of the cells, spores and Extra Polymeric Substratefibers intermingled on the surface of the coupon. A jelly-like substance could be observed among the corrosion products,which was attributed to biofilm produced EPS [25]. Elongated rod-shaped cells and round-shaped with round head cellsoccupied a small volume fraction compared to the precipitated corrosion products and EPS. The sizes of the cells range from1 to 5 lm. EPS and corrosion products have been reported to occupy 75–95% of biofilm volume, while 5–25% is occupied bythe metabolizing cells [25].

Fig. 5 presents the details of XRD data corresponding to the phases present in the corrosion product sample collected fromthe system exposed to biotic conditions for 45 days. The XRD pattern confirmed the formation of iron oxide compounds thatinclude iron, Hematite (Fe2O3), magnetite and iron (II) hydroxide Fe(OH)2.

F.M. AlAbbas et al. / Engineering Failure Analysis 33 (2013) 222–235 225

Fig. 6 displays FESEM images of the steel surface exposed to the abiotic system after cleaning. As shown in Fig. 6, the steelsurface exhibited aggressive pitting colonies. In comparison, Fig. 7 displays FESEM images of the steel surface exposed to thebiotic system, which reveal minor pitting as is evident with the visible polishing marks. These images suggest the protectionbehavior prompted by the activities of the IRB consortium in the case of the biotic system.

The compositions and structures of the corrosion products developed on the steel surfaces in abiotic and biotic systemssuggest the following reactions occurred:

3.2.1. For abiotic systemThe anodic reaction includes the iron dissolution to produce ferrous ions (Fe2+) as per the following reaction:

2Fe0 ! 2Fe2# # 4e" $1%

In neutral aerobic conditions, the total cathodic reactions include the reduction of dissolved oxygen and water dissocia-tion, reactions below [25,26]

1=2O2 #H2O# 2e" ! 2OH" $2%

Fig. 3. FESEM for the API 5L !52 carbon steel surface: (A) FESEM image of carbon steel exposed to abiotic system at 500!. (B). FESEM Image of carbon steelexposed to biotic system at 500!. (C) EDS spectra from abiotic (D) EDS spectra for biotic system.

Table 2Comparison of EDS analysis corresponding to the abiotic and the biotic systems.

Element (Wt%) C O Na Cl Si Fe N P Total

Biotic system 15.86 30.52 1.19 1.30 3.34 33.08 2.26 12.45 100Abiotic system 6.78 35.58 1.34 1.66 _ 54.63 _ _ 100

226 F.M. AlAbbas et al. / Engineering Failure Analysis 33 (2013) 222–235

2H2O# 2e" ! H2 # 2OH" $3%

Since the enriched growth medium solution contains sodium chloride (10 g/L), the following additional cathodic reaction(4) is also expected [26]:

Fe2# # 2Cl" ! Fe$Cl%2 $4%

Fig. 4. FESEM image for the biofilm developed on the API !52 exposed to the IRB containing medium !1000, !2000.

20 40 60 80 1000

200

400

600

800

1000

1200

1400

!" •

""!

!

Fe" Fe2O3! Fe3O4• FeOOH

Inte

nsity

(CPS

)

2# (Degree)

Fig. 5. XRD spectra for the system under biotic conditions.

F.M. AlAbbas et al. / Engineering Failure Analysis 33 (2013) 222–235 227

The ferrous ions will react with hydroxide and chloride ions to form iron hydroxide and iron chloride as corrosion prod-ucts and results in the following overall reaction (5):

2Fe2# # 2OH" # 2Cl" ! Fe$OH%2 # Fe$Cl%2 $5%

Therefore, in the abiotic system, the corrosion products are primarily composed of iron hydroxide and iron chlorides assuggested by the elemental EDS results, Fig. 3c. The iron dissolution is an activation control process, while the cathodic oxy-gen reduction is controlled by the diffusion process. The diffusion control process implies that the corrosion current density(icorr) equals the limiting current density (iL) [26]. In our experimental conditions, the limiting current density increased dueto the stirring effect [26]. Changes in the limiting current density can cause changes in the electrochemical kinetics of thesystem. In addition, the presence of sodium chloride along with dissolved oxygen in the growth medium, enhanced by stir-ring effects, explains the aggressive pitting under abiotic conditions. This mechanism is known to induce localized anodes onthe surface, subjecting the material to localized corrosion processes. This process is similar to the differential aeration cell,well documented in corrosion literature [26]. Once initiated, localized corrosion processes, such as pitting, are autocatalyticin nature [27]. Classic autocatalytic pitting mechanisms include film rupture, dissolution, hydrolysis, acidification, and chlo-ride migration [27]. A schematic representation that summarizes the corrosion mechanism under abiotic conditions isshown in Fig. 8a.

3.2.2. For biotic systemThe presence of an IRB consortium in abiotic system alters the corrosion behavior due to the induced effects by the IRB

bacterial metabolic activities. S. oneidensis are facultative anaerobic IRB which use oxygen as a terminal electron acceptor inaerobic conditions and utilize insoluble ferric ion as an electron acceptor in the case of anaerobic conditions [1,5,6,14].S. oneidensis form a biofilm on the metal surface, utilize tryptone, the carbon source, as electron donor and scavenge adjacentoxygen molecules via their aerobic respiration [5,6]. A schematic representation that summarizes the corrosion mechanismunder biotic conditions is shown in Fig. 8b. As oxygen is depleted, the bacteria turn to Fe3+ anaerobic respiration and

Fig. 6. FESEM analysis for the API 5L !52 carbon steel surface after cleaning for the system under abiotic conditions exposure at 200! and 500!.

Fig. 7. FESEM analysis for the API 5L !52 carbon steel surface after cleaning for the system under biotic conditions at 200! and 500!.

228 F.M. AlAbbas et al. / Engineering Failure Analysis 33 (2013) 222–235

produced Fe2+ ions diffuse into the bulk fluid.[1,5,6,14] The chemical and electrochemical processes in the case of a bioticsystem include iron dissolution, reaction (1) and the followings reactions [1,6,14,26]:

& IRB metabolic activities include aerobic depletion of oxygen (reaction 6) and anaerobic consumption of ferric ions (reac-tion 7) as electron acceptors

CxHyOz # O2 ! H2O# CO2 $6%

Fe3# # 1e" ! Fe2# $7%

& Formation of different iron oxides2Fe2# # 4OH" ! 2Fe$OH%2 $8%

4e" # 2Fe2# # 3H2O! Fe2O3 # 3H2 $9%

3Fe$OH%2 ! Fe3O4 # 2H2O#H2 $10%

In the case of a biotic system, the protective behavior of the IRB consortium might be explained by the following simul-taneous mechanisms:

Fig. 8. Schematic representation of Corrosion mechanism by IRB under (a) abiotic system and (b) biotic system.

F.M. AlAbbas et al. / Engineering Failure Analysis 33 (2013) 222–235 229

(a) The increased consumption of oxygen by S. oneidensis sp. respiration decreases the limiting current density resultingin lower corrosion rate. Dubiel et al. [8] reported that mutants of S. oneidensis sp. inhibit corrosion on stainless steelsurfaces in a medium containing yeast extract and peptone. They proposed that S. oneidensis sp. could be used as apotential method to control corrosion in pipeline systems [8]. S. oneidensis sp. may colonize the metal surface and con-sume oxygen molecules adjacent to the metal surface by aerobic respiration.

(b) As oxygen is depleted, the bacteria turn to Fe3+ anaerobic respiration and the produced Fe2+ ions diffuse into the bulkfluid [5]. This process will create a chemical shield that reduces oxygen diffusion, which in turn, inhibits the cathodicreaction due to lower oxygen availability. By the electrochemical reaction, Fe2+ is oxidized to Fe3+ and is again reducedby bacterial respiration. Consequently, corrosion is inhibited [1], [6].

(c) Another mechanism that might be responsible for corrosion inhibition is related to the generation of a cathodic pro-tection current by the bacterial consortium. Fig. 9 reveals that some bacterial cells possess a nano-wire structure. Itis possible these bacteria use nanowires to transfer cathodic current to the steel surface and, in turn, protect againstcorrosion. It was reported that S. oneidensis sp. produced electrically conductive pilus-like appendages called bacte-rial nanowires that can transfer electrons to the electrode [28]. The capacity of S. oneidensis sp. to transfer electronsfrom organic sources to electrodes without intervening catalysts serves as the basis for electricity production inmicrobial fuel cells (MFC) [28]. This postulation is further supported with our finding of the presence of Brevibacillussp. as part of this isolated aerobic consortium. Brevibacillus sp. have been investigated in microbial MFC for theircapacity to conduct electrons [23]. Interestingly, previous studies show that the ability of Brevibacillus sp. to conductelectrons was enhanced in the presence of Pseudomonas sp. The metabolites produced by Pseudomonas sp. enableBrevibacillus sp. to achieve extracellular electron transfer [23]. Based on these facts, it is possible that Brevibacillussp. and S. oneidensis sp., through their complimentary metabolisms, produce cathodic currents that provide extracorrosion inhibition.

3.3. Electrochemical evaluations

To gain more insight about evolution and kinetics of the corrosion process for both the abiotic and biotic systems, elec-trochemical tests were performed including open circuit potential, polarization resistance and electrical impedance spec-troscopy. These tests are considered nondestructive methods and provide in situ monitoring of the corrosion processwithout affecting biological aspects and solution chemistry.[8]

3.4. Open circuit potential/polarization resistance

The open circuit potential variations for the biotic and abiotic systems are shown in Fig. 10. The Ecorr as a function of timedata revealed that in the biotic system, the corrosion potential decreased to active value ("720 mV/SCE) at 50 h followed bya substantial shift towards noble values ("580 mV/SCE) at 100 h, which then remained stable throughout the period of expo-sure. On the other hand, in the abiotic system, there was a notable increase of the Ecorr, which then remained more or lesssteady at approximately "680 mV/SCE, Fig. 10.

The polarization resistance variations for the biotic and abiotic systems are shown in Fig. 11a. The Rp as a function of timedata revealed that in the biotic system, there was a substantial increase of polarization resistance (Rp) to '7000 X cm2 fol-lowed by a decrease to '6500 X cm2 at 150 h and then it remained stable at '5000 X cm2. In striking contrast, the Rp of theabiotic system shows a drop from '2000 X cm2 to '500 X cm2 at 200 h and then remains stable over the remaining period,

BinaryFission

EPS

Nanowiresand fibers

Fig. 9. FESEM image for the biofilm developed with arrows point to bacterial cells with nanowire structures.

230 F.M. AlAbbas et al. / Engineering Failure Analysis 33 (2013) 222–235

Fig. 11a. When comparing the two Rp trends for the biotic and abiotic system, the Rp values for the abiotic system are sig-nificantly lower than those for the biotic system. This observation suggests that the corrosion rate is substantially higherunder an abiotic system.

Polarization resistance is inversely proportional to corrosion current density [26,29]. The relation used to determine cor-rosion current density, icorr, using the polarization resistance method, can be derived from mixed potential theory using thefollowing relationship:

Rp(X cm2) * $ba + bc=$2:3icorr$ba # bc%% $11%

where ba and bc are the Tafel slops for the anodic and cathodic reactions, respectively.The Rp trend confirms the corrosion inhibition by the IRB consortium as it reveals that current density decreased with

time. As the bacteria reached the death phase at around 150 h, their activity decreased; subsequently, their capacity for cor-rosion inhibition was less. This behavior might explain the decrease of Rp at 200 h, Fig. 11a.

The corrosion rate plots over time for the biotic and abiotic systems are shown in Fig. 11b. The corrosion rate for the abi-otic medium reached a maximum value of 20 mpy after 200 h, whereas the corrosion rate for the biotic system was around 4mpy over the experimental period.

The drop of Ecorr to active value in the biotic system has been related to the growth and aerobic metabolic activities of theIRB consortium. IRB colonize the surface and consume the adjacent oxygen molecules, reaction 6, resulting in a decrease of

0 50 100 150 200 250-720

-700

-680

-660

-640

-620

-600

-580

-560

OC

P (m

V/SC

E)

Time (hrs)

Abiotic Biotic

Fig. 10. Open Circuit Potential (OCP) variations under biotic and abiotic conditions.

0

1000

2000

3000

4000

5000

6000

7000

8000

Rp

($.c

m2 )

Time (hrs)

Abiotic Biotic

0 50 100 150 200 0 50 100 150 2000

2

4

6

8

10

12

14

16

18

20

22C

orro

sion

rate

(mpy

)

Time (hrs)

Abiotic Biotic

A B

Fig. 11. (A) Polarization resistance and (B) Corrosion rate variations for both biotic and abiotic systems.

F.M. AlAbbas et al. / Engineering Failure Analysis 33 (2013) 222–235 231

the limiting current density (iL) and decrease in the Ecorr. However, when oxygen is depleted, IRB will switch to anaerobicrespiration by reducing the ferric ions (Fe3+) to ferrous ions (Fe2+), reaction 7, and in turn, increase the ferrous ions’ concen-tration in the solution [7,8]. This mechanism shifts the reversible potential (Ecorr) to a more noble value and decreases thecorrosion current density (icorr) as evident in Figs. 10 and 11a. These observations are consistent with the different investi-gations’ findings when mild steel and stainless steel were exposed to mutants of S. oneidensis [8,30].

On the other hand, Nagiub and Mansfeld [30] have shown a decrease of Ecorr and icorr for brass and aluminum alloy 2024when they were exposed to two strains of Shewanella sp. that have been attributed to aerobic metabolic activities. There is adifference in noble direction of approximately 100 mV/SCE between the biotic and abiotic systems. This positive shift in Ecorr

is known as ennoblement. Ennoblement has been reported for different alloys exposed to IRB strains [3,8]. Ennoblement hasbeen cited as probably the most notable phenomenon in many MIC investigations [3]. The microbial colonization, biofilmformation and iron reduction of the IRB consortium play a significant role in the ennoblement phenomena [31].

The Ecorr, Rp and corrosion rate data confirm the corrosion inhibition properties of the IRB in accordance with the mech-anism proposed above. There is no single mechanism to explain the inhibition process; however, the best proposed hypoth-esis in accordance with the literature and our findings has already been stated above. Furthermore, to support ourhypothesis, the dominant protection phase by the IRB was through the oxygen removal during the first 50 h (shown by adecrease in Ecorr and an increase in Rp), followed by the anaerobic reduction of the ferric ions to the ferrous ions for the next150 h with the development of a protection shield which further prevents diffusion of oxygen and chloride to the surface(shown by an increase in Ecorr and an increase in Rp) and followed by bacterial growth limitation in the stationery or declinephase. The role of the other bacterial species, Brevibacillus sp., working in synergism cannot be excluded.

3.5. Electrical impedance spectroscopy (EIS)

Fig. 12a displays the Nyquist plots for the carbon steel coupon exposed to the abiotic system over time. The steady statewas reached at 120 h. At low frequencies (LF), shown in Fig. 12a, the magnitude of the capacitive loop represented by thesemicircle diameter decreased with time from '2500 X cm2 to 500 X cm2. These LF magnitudes represent the change incharge transfer resistance (Rct) that describes the evolution of the anodic reaction [25]. It is apparent from the Nyquist plotsthat there is a significant decrease in the low frequency limit of the impedance for the system. This observation fits well withthe significant increasing corrosion rates observed in the Rp data. The decrease of Rct with time indicates an increase in cor-rosion rate, possibly due to the effect of oxygen and chloride in the medium; and subsequent formation of a mixed layer ofiron hydroxides, iron chloride and carbon-based compounds on the electrode surface [8,25]. In this case, the presence of oxy-gen and chloride in the system enhances the cathodic reactions that drive the anodic dissolution of the metal. The formationof a corrosion product layer was confirmed by the two capacitive loop impedance spectra at 72 h and phase angle spectra(Fig. 12b) that shows two time constants at 10 Hz and 0.1 Hz medium frequencies.

Based on the minimum deviation between the measured and fitted data, the impedance spectra were modeled with anequivalent electrical circuit that has two time constants as shown in Fig. 13.

The circuit includes: (1) resistance Rs considered as solution resistance, (2) parallel combination of a charge transfer resis-tance (Rct) and constant phase element (CPEdl) associated with steel surface double layer capacitance and (3) another porousparallel combination of a resistance, Rpo and constant phase element, CPEpo associated with the formation of a heterogeneouslayer composed of corrosion products along with other compounds deposited from the growth media. In general, a CPE is usedinstead of a capacitor to compensate for the ideal behavior. The impedance of CPE is defined by the following equation [32,33]:

0 1000 2000 30000

-1000

-2000

-3000

Z"($

.cm

2 )

Z ($.cm2)

24 hrs

72 hrs

120 hrs

168 hrs

216 hrs

10-2 10-1 100 101 102 103 104 105

20

0

-20

-40

-60

-80

Phas

e (D

egre

es)

Frequency (Hz)

24 hrs 72 hrs 120 hrs 168 hrs 216 hrs

A B

Fig. 12. EIS data for abiotic culture medium; (A) Nyquist Plots (B) Phase angle plots.

232 F.M. AlAbbas et al. / Engineering Failure Analysis 33 (2013) 222–235

ZCPE * $CPE%"1$jx%"a $12%

When the carbon steel was exposed to the biotic system, the EIS spectra varied significantly with exposure time as shownin Fig. 14a. The low frequency (LF) magnitude, represented by the semicircle diameter, significantly increased with time,indicating an increase in the charge transfer resistance (Rct). The low frequency (10"1 Hz) response presented in the phasediagram, Fig. 14b, shows two time constants at 120 h, which indicate the formation of a heterogeneous biofilm layer basedon a mixture of inorganic/organic compounds [2,31,32,33]. It is possible that the IRB metabolism of ferric ions develops aprotective shield that acts as a barrier to corrosion. These EIS spectra for the biotic system provide further confirmationof the inhibitory actions of the IRB through the mechanisms described in the previous sections.

The EIS spectra suggests the interfacial evolution under the biotic system followed the below mentioned steps;

i. Before 120 h, the medium frequency range showed a one time constant behavior which can be attributed to the for-mation of a surface layer comprising of a mixture of corrosion products, mainly ferric hydroxide and organiccompounds.

ii. After 120 h, the mixed layer of corrosion products and biofilm was established as evident by the two time constantbehavior in the phase angle curves.

iii. After 168 h, due to the limitation in the bacterial nutrition and accumulation of bacterial by-products of metabolismthe proliferation of IRB decreased which compromised the integrity of the protective layer there by showing anupward increase in the corrosion rate. It is also evident by the change in the shape of nyquist curve from a semicircleto a straight line (Fig. 14a)

The impedance spectra were modeled with an equivalent electrical circuit that has two time constants as shown inFig. 15. The circuit consists of: (1) solution resistance, Rs, (2) parallel combination of biofilm capacitance, CPEbf and resis-tance, Rbf and (3) parallel combination of double layer capacitance, CPEdl and charge transfer resistance, Rct for the steelsurface.

Fig. 13. Proposed equivalent circuit for the abiotic system.

0 1500 3000 4500 60000

-1500

-3000

-4500

-6000 24 hrs

72 hrs

120 hrs

168 hrs

216 hrs

Z"($

.cm

2 )

Z ($.cm2)10-2 10-1 100 101 102 103 104 105

0

-20

-40

-60

-80 24 hrs

72 hrs

120 hrs

168 hrs

216 hrs

Phas

e (D

egre

es)

Frequency (Hz)

A B

Fig. 14. EIS data for inoculated biotic medium; (a) Nyquist Plots (b) Phase angle plots.

F.M. AlAbbas et al. / Engineering Failure Analysis 33 (2013) 222–235 233

4. Conclusions

This study has investigated the microbiologically influenced corrosion of API 5L !52 carbon steel coupons exposed to awild type of iron-reducing consortium cultivated from a sour oil well produced water. 16S rRNA gene sequence analysis indi-cated that the mixed bacterial culture consortium contained two phylotypes that are close to members of the Proteobacteria(S. oneidensis sp.) and Firmicutes (Brevibacillus sp.). For the biotic system, X-ray diffraction confirmed the presence of ironoxide compounds that include iron, Hematite (Fe2O3), magnetite (Fe3O4) and iron (II) hydroxide Fe(OH)2. The IRB consortiumexhibited inhibitory action on the corrosion process. The maximum corrosion rate in the biotic system was '4 mpy while itwas '20 mpy in the abiotic system. The steel surface exposed to the abiotic system showed extensive pitting, whereas lesspitting was found on the steel surface in contact with the biotic system as evident by the visible polishing marks. Electro-chemical evaluation, (OCP, Rp and EIS) confirm the inhibition effects of the wild IRB consortium. Our research suggests thatthe IRB consortium could be used as a potential source to combat corrosion especially in an aerobic environment.

Acknowledgements

The authors acknowledge and appreciate the Saudi Aramco and Inspection Department Management for their continualsupport for this project.

References

[1] Javaherdashti R. Microbiologically influenced corrosion: an engineering insight. London: Springer-Verlag; 2008.[2] Alabbas FM, Gavanluei AB, Spear JR, et al, Effects of sulfate reducing bacteria on the corrosion of X-65 pipeline carbon steel. In: Proceeding NACE 2012

conference, Paper # C2012-0001140; 2011.[3] Little BJ, Lee JS. Microbiologically influenced corrosion. Hoboken, NJ, USA: John Wiley & Sons Inc.; 2007. p. 1–50.[4] Walch M. Corrosion, microbial. In: Lederberg J, editor. Encyclopaedia of microbiology, vol. 1. New York: Academic Press; 1992. p. 585–91.[5] Nealson KH, Saffarini D. Iron and manganese in anaerobic respiration: environmental significance, physiology, and regulation. Annu Rev Microbiol

1994;48:311–43.[6] Herrera Lk, Videla HA. Role of iron-reducing bacteria in corrosion and protection of carbon steel. Int Biodeterior Biodegr 2009;63:891–5.[7] Videla HA, Herrera LK. Understanding microbial inhibition of corrosion. A comprehensive overview. Int Biodeterior Biodegr 2009;63:896–900.[8] Dubiel M, Hsu CH, Chien CC, Mansfield F, Newman DK. Microbial iron respiration can protect steel from corrosion. Appl Environ Microbiol

2002;68:1440–5.[9] Jayaraman JC, Earthman TK. Corrosion inhibition by aerobic biofilms on SAE 1018 steel. Appl Microbiol Biotechnol 1997;47:62–8.

[10] Videla HA. Corrosion inhibition by bacteria. In: Manual of biocorrosion. Boca Raton, Florida: CRC Lewis Publishers; 1996. p. 121–35.[11] Black JP, Ford TE, Mitchell R. Corrosion behavior of metal binding exopolymers from iron and manganese-depositing bacteria. In: Corrosion

88. Houston, Texas: National Association of Corrosion Engineers; 1988.[12] Javaherdashti R, Singh RK, Panter C, Pereloma EV. Microbiologically assisted stress corrosion cracking of carbon steel in mixed and pure cultures of

sulfate reducing bacteria. Int Biodeterior Biodegr 2006;58:27–35.[13] Chongdar S, Gunasekaran G, Kumar P. Corrosion inhibition of mild steel by aerobic biofilm. Electrochimica Acta 2005;50(24,30):4655–65.[14] Videla HA. Corrosion inhibition in the presence of microbial corrosion. CORROSION/96 NACE International, Houston paper no. 223; 1996.[15] Little B, Wagner P, Hart K, Ray R, Lavoie D, Nealson K, et al. The role of metal-reducing bacteria in microbiologically influenced corrosion. CORROSION/

97 paper no. 215. NACE International, Houston, TX; 1997.[16] NACE Standard TM0194, 2004. Field monitoring of bacterial growth in oil and gas systems. Houston, Texas: Published by NACE.[17] Sahl W, Jason F, Nathaniel HJ, Kirk W, David WC, Spear JR. Novel microbial diversity retrieved by autonomous robotic exploration of the world’s

deepest vertical phreatic sinkhole. Astrobiology 2010;10(2):201–13.[18] Frank D. XplorSeq: a software environment for integrated management and phylogenetic analysis of metagenomic sequence data. BMC Bioinformatics

2008;9:9–420.[19] ASTM G1-03. Standard practice for preparing, cleaning and evaluating corrosion test specimens. ASTM, Philadelphia; 2009. p. 17–23.[20] Lutterbach MT, Contador L, Oliveira A, Galvão M. Iron sulfide production by Shewanella strain isolated from black powder. In: Proceeding NACE 2009

conference, Paper # 09391; 2009.[21] Youssef N, Elshahedand MS, McInerney MJ. Microbial processes in oil fields: culprits, problems, and opportunities. Adv Appl Microbiol

2009;66:141–251.[22] Aelterman P, Rabaey K, Pham HT, Boon N, Verstraete W. Continuous electricity generation at high voltages and currents using stacked microbial fuel

cells. Environ Sci Technol 2006;40:3388–94.[23] Pham TH, Boon N, Aelterman P, et al. Metabolites produced by pseudomonas sp. enable a gram positive bacterium to achieve extracellular electron

transfer. Appl Microbiol Biotechnol 2008;77:1119–29.[24] Shobhana C, Gunasekaran G, Kumar P. Corrosion inhibition of mild steel by aerobic biofilm. Electrochimica Acta 2005;50(24):4655–65.[25] Castaneda H, Benetton XD. SRB-biofilm influenced in active corrosion sites formed at the steel-electrolyte interface when exposed to artificial seawater

conditions. Corros Sci 2008;50:1169–83.

Fig. 15. Proposed equivalent circuit for the biotic system.

234 F.M. AlAbbas et al. / Engineering Failure Analysis 33 (2013) 222–235

[26] Jones DA. Principles and prevention of corrosion. 2nd ed. NJ, USA: Simon & Schuster/Viacom Company; 1996.[27] Fontana Mars G, Greene Norbert D. Corrosion engineering. New York: McGraw-Hill Book Company; 1967.[28] Gorby, Svetlana YAY, McLean JS, et al. Electrically conductive bacterial nanowires produced by Shewanella oneidensis strain MR-1 and other

microorganisms. PNAS 103; 2006. p. 11358–63.[29] Bhola SM, Bhola R, Jain L, Mishra B, Olson DL. Corrosion behavior of mild carbon steel in ethanolic solutions. J Mater Eng Perform 2011;20(3):409–16.[30] Nagiub L, Mansfeld F. Evaluation of microbiologically influenced corrosion inhibition (MICI) with EIS and ENA. Electrochimica Acta 2002;47:2319–33.[31] Liu J, Liang X, Li S. Effect of sulphate-reducing bacteria on the electrochemical impedance spectroscopy characteristics of 1 Cr18Ni9Ti. J Univ Sci

Technol Beijing 2007;14(5):425–30.[32] Bhola SM, Bhola R, Mishra B, Olson DL. Electrochemical impedance spectroscopic characterization of the oxide film formed over low modulus Ti–

35.5Nb–7.3Zr–5.7Ta alloy in phosphate buffer saline at various potentials. J Mater Sci 2010;45(22):6179–86.[33] Kuang F, Wang J, Yana L, Zhang D. Effects of sulfate-reducing bacteria on the corrosion behavior of carbon steel. Electrochimica Acta. 2007;52:6084–8.

F.M. AlAbbas et al. / Engineering Failure Analysis 33 (2013) 222–235 235